CN115660728A

CN115660728A - Air ticket sales order prediction method and device, electronic equipment and storage medium

Info

Publication number: CN115660728A
Application number: CN202211241700.0A
Authority: CN
Inventors: 石路路; 赵英廷; 孟平; 史超
Original assignee: Nanjing Yibo Software Technology Co ltd
Current assignee: Nanjing Yibo Software Technology Co ltd
Priority date: 2022-10-11
Filing date: 2022-10-11
Publication date: 2023-01-31
Anticipated expiration: 2042-10-11
Also published as: CN115660728B

Abstract

The application relates to a method and a device for predicting an air ticket sales order, electronic equipment and a storage medium. Wherein, the method comprises the following steps: obtaining air ticket order history data; obtaining target prediction data according to the historical target data and at least one characteristic; the historical target data comprises first order quantity information of the air ticket sales orders in a first time period in the air ticket order historical data; at least one characteristic is constructed according to the travel information of the air ticket sales order in the air ticket order historical data; the target forecast data comprises second order quantity information of the air ticket sales orders in a specified time period in at least one airline department commission-returning protocol; determining whether to recommend a protocol flight to the user according to the target prediction data; wherein the agreement flight comprises at least one flight corresponding to the flight refund agreement. The accuracy of the prediction result can be improved through the method and the device, the navigation department can reasonably complete the commission returning protocol, the completion rate of the commission returning protocol of the navigation department is improved, and the income brought by the commission returning protocol of the navigation department is improved.

Description

Ticket sales order prediction method and device, electronic equipment and storage medium

Technical Field

The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for predicting an air ticket sales order, an electronic device, and a storage medium.

Background

Travel Management Companies (TMC) can help enterprises to perform overall planning and overall execution monitoring on Travel activities with the assistance of a professional Travel Management service team, optimize Travel Management processes and policies, and integrally purchase resources, so that Travel cost of enterprises is reduced and trip efficiency of employees is improved on the premise of not influencing business development and trip experience. TMC provides a travel booking platform for enterprise employees to book traffic and accommodation in travel. The traffic comprises air tickets, train tickets and used cars, wherein the air ticket resources are mainly sourced from navigation/suppliers, and TMC companies sign a series of contracts of flight refreshing, agreement price, commission return and the like with the navigation/suppliers according to the air ticket travel volume of service enterprises in the calendar year.

The department of aviation commissions back agreement will agree that: under the conditions of a specified time range, a specified round-trip city, a specified flight and the like, when the total fare or the flight segment quantity meets a specified quantity, the specified condition is given a commission back; conversely, if the specified amount is not met, the commission back is 0. Therefore, how to ensure that more airline department commission returns are obtained becomes a business problem which needs to be solved urgently.

Disclosure of Invention

In view of this, an air ticket sales order prediction method, an air ticket sales order prediction apparatus, an electronic device, and a storage medium are provided.

In a first aspect, an embodiment of the present application provides a method for predicting an air ticket sales order, where the method includes: obtaining air ticket order history data; obtaining target prediction data according to the historical target data and at least one characteristic; the historical target data comprises first order quantity information of the air ticket sales orders in a first time period in the air ticket order historical data; the at least one characteristic is constructed according to the travel information of the air ticket sales order in the air ticket order historical data; the target forecast data comprises second order quantity information of the air ticket sales orders in a specified time period in at least one airline department commission-returning protocol; determining whether to recommend a protocol flight to a user according to the target prediction data; wherein the agreement flight comprises a flight corresponding to the at least one airline commission return agreement.

Based on the technical scheme, target prediction data of the air ticket sales order are obtained according to historical target data and at least one characteristic of the air ticket sales order, and the at least one characteristic is constructed according to the travel information of the air ticket sales order in the historical data of the air ticket order, so that the accuracy of a prediction result is improved; whether the scheduled flight in the airline department commission returning protocol is recommended to the user is determined according to the target prediction data of the air ticket sales order, so that the airline department commission returning protocol can be reasonably completed, the completion rate of the airline department commission returning protocol is improved, and the income brought by the airline department commission returning protocol is improved.

According to the first aspect, in a first possible implementation manner of the first aspect, the method further includes: obtaining a recommendation index of at least one agreement flight according to the user portrait and the agreement flight portrait under the condition of determining that the agreement flight is recommended to the user; the user portrait is constructed according to historical order information of a user; the agreement flight portrait is constructed according to the information of the agreement flight; and recommending agreement flights to the user according to the recommendation index.

Based on the technical scheme, under the condition that the agreement flight is determined to be recommended to the user, the recommendation index of the agreement flight is calculated according to the user portrait and the agreement flight portrait, and the agreement flight is recommended to the user according to the recommendation index, so that the ordering rate of the agreement flight ticket by the user is improved, the momentum of the airline company commission returning protocol which does not complete the target can be realized, the completion rate of the airline company commission returning protocol is improved, and the income brought by the airline company commission returning protocol is improved.

According to the first aspect or the first possible implementation manner of the first aspect, in a second possible implementation manner of the first aspect, the obtaining target prediction data according to historical target data and at least one feature includes: obtaining a historical time sequence according to the historical target data; inputting the historical time sequence and the at least one characteristic into a preset model to obtain the target prediction data; the preset model is obtained by training based on an STL time sequence decomposition algorithm.

Based on the technical scheme, the historical time sequence and the characteristics are input into the preset model obtained by training based on the STL algorithm to obtain target prediction data, and the extraction method of the period item sequence and the trend item sequence in the STL algorithm can be optimized, so that the traditional STL algorithm is improved, the problem of dependence on local data is solved, a more accurate time sequence decomposition result can be obtained, the accuracy of the prediction result can be improved, and more accurate target prediction data can be obtained.

According to a second possible implementation manner of the first aspect, in a third possible implementation manner of the first aspect, the preset model includes a first submodel, a second submodel, and a third submodel; the first sub model, the second sub model and the third sub model are obtained by training based on a regression algorithm; the inputting the historical time series and the at least one feature into a preset model to obtain the target prediction data includes: obtaining an input sequence according to the historical time sequence and the at least one characteristic; acquiring a trend item sequence in the (k-1) th iteration, and acquiring a first intermediate sequence according to the input sequence and the trend item sequence in the (k-1) th iteration; wherein the initial value of k is 1, and the initial value of the trend item sequence is 0; inputting the first intermediate sequence into the first submodel to obtain a second intermediate sequence; obtaining a third intermediate sequence according to the second intermediate sequence; inputting the third intermediate sequence into the second submodel to obtain a fourth intermediate sequence; obtaining a periodic item sequence and a fifth intermediate sequence in the kth iteration according to the input sequence, the second intermediate sequence and the fourth intermediate sequence; inputting the fifth intermediate sequence into the third submodel to obtain a trend term sequence in the kth iteration; judging whether the periodic item sequence in the kth iteration and the trend item sequence in the kth iteration are converged, stopping the iteration under the condition that the periodic item sequence in the kth iteration and the trend item sequence in the kth iteration are converged, determining the periodic item sequence in the latest iteration as the periodic item sequence corresponding to the input sequence, and determining the trend item sequence in the latest iteration as the trend item sequence corresponding to the input sequence; otherwise, adding 1 to the value of k, and returning to execute the step of acquiring the trend item sequence in the (k-1) th iteration; and obtaining the target prediction data according to the period item sequence corresponding to the input sequence and the trend item sequence corresponding to the input sequence.

Based on the technical scheme, the historical time sequence and the characteristics are input into a preset model obtained by training based on an STL algorithm to obtain target prediction data, and the extraction method of the period item sequence and the trend item sequence is optimized, so that the traditional STL algorithm is improved, the problem of dependence on local data is solved, the sequence information is accurately decomposed and divided and controlled by using a first sub-model, a second sub-model and a third sub-model which are obtained by calculation based on a regression algorithm, more accurate time sequence decomposition is realized, the accuracy of a prediction result can be improved, and more accurate target prediction data can be obtained.

According to a third possible implementation manner of the first aspect, in a fourth possible implementation manner of the first aspect, the obtaining a periodic term sequence and a fifth intermediate sequence in a kth iteration according to the input sequence, the second intermediate sequence, and the fourth intermediate sequence includes: subtracting the fourth intermediate sequence from the second intermediate sequence to obtain a periodic term sequence in the k iteration; and subtracting the periodic item sequence in the k iteration from the input sequence to obtain the fifth intermediate sequence.

Based on the technical scheme, the historical time sequence and the characteristics form the input of the trend item sequence and the periodic item sequence in the time sequence, the extraction method of the periodic item sequence and the trend item sequence is optimized, and the more accurate periodic item sequence and trend item sequence can be obtained, so that the accuracy of the prediction result can be improved, and more accurate target prediction data can be obtained.

In a fifth possible implementation form of the first aspect according to the first aspect or the above-mentioned various possible implementation forms of the first aspect, the objective prediction data further includes third order quantity information of the airline ticket sales order in a second time period corresponding to the at least one airline commission return agreement; and the third order quantity information of the air ticket sales orders in the second time period corresponding to the at least one airline commission returning agreement is used for evaluating the airline commission returning agreement in the second time period.

Based on the technical scheme, by predicting the order quantity information of the airline department commission returning protocol in a period of time in the future, reference can be provided for the establishment of the airline department commission returning protocol in the next stage, and the establishment of a more reasonable airline department commission returning protocol is facilitated, so that the completion rate of the airline department commission returning protocol is improved, and the income brought by the airline department commission returning protocol is improved.

In a sixth possible implementation manner of the first aspect, according to the first aspect or the above-mentioned various possible implementation manners of the first aspect, the first order quantity information and the second order quantity information include one or more of a total fare, a leg amount, and a commission return amount.

Based on the technical scheme, the total ticket face price, the flight segment amount and the commission return amount of the air ticket sales order which meet the requirements of the airline company commission return agreement in the air ticket order historical data are used for predicting the total ticket face price, the flight segment amount and the commission return amount of the airline company commission return agreement in a period of time in the future, and the predicted target completion rate of the airline company commission return agreement can be calculated, so that whether the impulse is carried out on the airline company commission return agreement which does not finish the target is determined, the airline company commission return agreement is reasonably completed, the completion rate of the airline company commission return agreement is improved, and the income brought by the airline company commission return agreement is improved.

In a seventh possible implementation manner of the first aspect, according to the first aspect or the above-mentioned various possible implementation manners of the first aspect, the itinerary information includes one or more of date information, holiday information, city information, business application information.

Based on the technical scheme, the holiday characteristics can be constructed according to the date information and the holiday information of the air ticket sales order in the air ticket order historical data, the business trip application number characteristics can be constructed according to the city information and the business trip application information, the holiday characteristics and the business trip application number characteristics are used for predicting target prediction data, and the accuracy of prediction results is improved.

In a second aspect, an embodiment of the present application provides an air ticket sales order prediction apparatus, including: the acquisition module is used for acquiring air ticket order history data; the prediction module is used for obtaining target prediction data according to the historical target data and at least one characteristic; the historical target data comprises first order quantity information of the air ticket sales orders in a first time period in the air ticket order historical data; the at least one characteristic is constructed according to the travel information of the air ticket sales order in the air ticket order historical data; the target forecast data comprises second order quantity information of the air ticket sales orders in a specified time period in at least one airline department commission-returning protocol; the recommendation module is used for determining whether to recommend a protocol flight to the user according to the target prediction data; wherein the agreement flight comprises a flight corresponding to the at least one airline commission return agreement.

Based on the technical scheme, target prediction data of the air ticket sales order are obtained according to historical target data and at least one characteristic of the air ticket sales order, and the at least one characteristic is constructed according to the travel information of the air ticket sales order in the historical data of the air ticket order, so that the accuracy of a prediction result is improved; whether the scheduled flight in the airline company commission returning protocol is recommended to the user is determined according to the target prediction data of the air ticket sales order, so that the airline company commission returning protocol is reasonably completed, the completion rate of the airline company commission returning protocol is increased, and the income brought by the airline company commission returning protocol is increased.

According to the second aspect, in a first possible implementation manner of the second aspect, the recommending module is further configured to: obtaining a recommendation index of at least one agreement flight according to the user portrait and the agreement flight portrait under the condition of determining that the agreement flight is recommended to the user; the user portrait is constructed according to historical order information of a user; the protocol flight portrait is constructed according to the information of the protocol flight; and recommending a protocol flight to the user according to the recommendation index.

In a second possible implementation manner of the second aspect, the prediction module is further configured to: obtaining a historical time sequence according to the historical target data; inputting the historical time sequence and the at least one characteristic into a preset model to obtain the target prediction data; and the preset model is obtained by training based on an STL time series decomposition algorithm.

According to a second possible implementation manner of the second aspect, in a third possible implementation manner of the second aspect, the preset model includes a first sub-model, a second sub-model, and a third sub-model; the first sub-model, the second sub-model and the third sub-model are obtained by training based on a regression algorithm; the prediction module is further configured to: obtaining an input sequence according to the historical time sequence and the at least one characteristic; acquiring a trend item sequence in the (k-1) th iteration, and acquiring a first intermediate sequence according to the input sequence and the trend item sequence in the (k-1) th iteration; wherein the initial value of k is 1, and the initial value of the trend item sequence is 0; inputting the first intermediate sequence into the first submodel to obtain a second intermediate sequence; obtaining a third intermediate sequence according to the second intermediate sequence; inputting the third intermediate sequence into the second submodel to obtain a fourth intermediate sequence; obtaining a periodic item sequence and a fifth intermediate sequence in the kth iteration according to the input sequence, the second intermediate sequence and the fourth intermediate sequence; inputting the fifth intermediate sequence into the third submodel to obtain a trend term sequence in the kth iteration; judging whether the periodic item sequence in the kth iteration and the trend item sequence in the kth iteration are converged, stopping the iteration under the condition that the periodic item sequence in the kth iteration and the trend item sequence in the kth iteration are converged, determining the periodic item sequence in the latest iteration as the periodic item sequence corresponding to the input sequence, and determining the trend item sequence in the latest iteration as the trend item sequence corresponding to the input sequence; otherwise, adding 1 to the value of k, and returning to execute the step of acquiring the trend item sequence in the (k-1) th iteration; and obtaining the target prediction data according to the period item sequence corresponding to the input sequence and the trend item sequence corresponding to the input sequence.

In a fourth possible implementation manner of the second aspect, according to the third possible implementation manner of the second aspect, the prediction module is further configured to: subtracting the fourth intermediate sequence from the second intermediate sequence to obtain a periodic term sequence in the kth iteration; and subtracting the periodic item sequence in the k iteration from the input sequence to obtain the fifth intermediate sequence.

In a fifth possible implementation form of the second aspect, according to the second aspect or the above-mentioned various possible implementation forms of the second aspect, the goal prediction data further includes third order quantity information of the airline ticket sales order in a second time period corresponding to the at least one airline commission return agreement; and the third order quantity information of the air ticket sales order in the second time period corresponding to the at least one airline turnback agreement is used for evaluating the airline turnback agreement in the second time period.

In a sixth possible implementation manner of the second aspect, according to the second aspect or the above-mentioned various possible implementation manners of the second aspect, the first order quantity information and the second order quantity information include one or more of a total fare, a leg quantity, and a commission return amount.

Based on the technical scheme, the total ticket face price, the flight segment amount and the commission return amount of the air ticket sales order which meet the requirement of the airline company commission return protocol in the historical data of the air ticket order are used for predicting the total ticket face price, the flight segment amount and the commission return amount of the airline company commission return protocol in a period of time in the future, and the predicted target completion rate of the airline company commission return protocol can be calculated, so that whether the impulse of the airline company commission return protocol which does not finish the target is needed or not is determined, the airline company commission return protocol is reasonably completed, the completion rate of the airline company commission return protocol is improved, and the income brought by the airline company commission return protocol is improved.

In a seventh possible implementation manner of the second aspect, according to the second aspect or the above various possible implementation manners of the second aspect, the itinerary information includes one or more of date information, holiday information, city information, business application information.

In a third aspect, an embodiment of the present application provides an electronic device, where the terminal device may perform the method for predicting an air ticket sales order in the first aspect or in one or more of the implementations of the first aspect.

In a fourth aspect, an embodiment of the present application provides a computer program product, which includes computer readable code or a non-transitory computer readable storage medium carrying computer readable code, and when the computer readable code is executed in an electronic device, a processor in the electronic device executes an air ticket sales order prediction method of one or more of the foregoing first aspect or multiple implementations of the first aspect.

In a fifth aspect, embodiments of the present application provide a computer-readable storage medium having stored thereon computer program instructions, which, when executed by a processor, implement the method for predicting an air ticket sales order of the first aspect or one or more of the various implementations of the first aspect.

Other features and aspects of the present application will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate exemplary embodiments, features, and aspects of the application and, together with the description, serve to explain the principles of the application.

Fig. 1 shows a flow diagram of an STL algorithm in accordance with an embodiment of the present application.

Fig. 2 shows a flow chart of a method for predicting a ticket sales order according to an embodiment of the present application.

FIG. 3 illustrates a schematic diagram of a user representation in accordance with an implementation of the present application.

FIG. 4 illustrates a schematic diagram of a user representation and a protocol flight representation in accordance with an implementation of the present application.

Fig. 5 shows a flow chart of a method for predicting a ticket sales order according to an embodiment of the present application.

FIG. 6 shows a flow chart for building a predictive model according to an embodiment of the present application.

FIG. 7 illustrates a schematic diagram of an airline ticket sales order prediction system according to an embodiment of the present application.

Fig. 8 (a) -8 (c) show schematic diagrams of an airline ticket sales order prediction system according to an embodiment of the present application.

Fig. 9 is a schematic diagram illustrating an air ticket sales order prediction method according to an embodiment of the present application and a prediction curve for predicting the total fare of the airline commission repayment agreement according to the STL algorithm.

Fig. 10 is a schematic diagram illustrating target lift rates after three airline department commission return protocols are pulsed by the airline ticket sales order prediction system according to an embodiment of the present application.

Fig. 11 is a block diagram illustrating an arrangement of an air ticket sales order prediction apparatus according to an embodiment of the present application.

Fig. 12 shows a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

Various exemplary embodiments, features and aspects of the present application will be described in detail below with reference to the accompanying drawings. In the drawings, like reference numbers can indicate functionally identical or similar elements. While the various aspects of the embodiments are presented in drawings, the drawings are not necessarily drawn to scale unless specifically indicated.

Reference throughout this specification to "one embodiment" or "some embodiments," or the like, means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the present application. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," or the like, in various places throughout this specification are not necessarily all referring to the same embodiment, but rather "one or more but not all embodiments" unless specifically stated otherwise. The terms "comprising," "including," "having," and variations thereof mean "including, but not limited to," unless expressly specified otherwise.

In the present application, "at least one" means one or more, "a plurality" means two or more. "and/or" describes the association relationship of the associated object, indicating that there may be three relationships, for example, a and/or B, which may indicate: including the presence of a alone, a and B together, and B alone, where a and B may be singular or plural. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of the singular or plural items. For example, at least one (one) of a, b, or c, may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be single or multiple.

Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the technical solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

The word "exemplary" is used exclusively herein to mean "serving as an example, embodiment, or illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments. Furthermore, in the following detailed description, numerous specific details are set forth in order to provide a better understanding of the present application. It will be understood by those skilled in the art that the present application may be practiced without some of these specific details.

The income management of air ticket is more ideal at international aviation department implementation effect, and generally aviation department implements the income management back, all can produce 3% -6% income promotion, and the main reason that international income management system can fall to the ground is as follows: (1) The international airline is a typical hub-type airline network, has few flights, few co-flying companies and relatively stable transportation structure; (2) Foreign travelers are usually reserved 2-3 weeks before the flight takes off, so that foreign forecasts of market demand (mid-term) are very accurate, meaning that price fighting can be accurately performed. The TMC company should also have its revenue management system as an agent of the airline department, but no system or product related to the airline department commission-returning agreement is found from the currently published data, and the upper layer application of the management data is not taken away.

In the related art, a Seasonal-Trend-based progress price (STL) algorithm based on local weighted regression may be used to predict the total fare and/or the amount of the flight segment of the ticket sales order within a specified time period in the flight return commission protocol, so as to provide a reference for how to reasonably complete the flight return commission protocol. The STL algorithm is a robust algorithm with good effect in time series decomposition, and can decompose a time series into a trend term sequence, a periodic term sequence and a residual term sequence. The STL algorithm has the following advantages: (1) The period item can change along with the change of time, and the change rate can be customized by a user; (2) the trend item may be customized by the user; (3) It is not sensitive to outliers, but may make local residuals large.

The STL algorithm performs local polynomial regression on the period term and the trend term through local weighted regression (locality). The Loess algorithm is a robust regression algorithm, is a common method for smoothing a two-dimensional scatter diagram, and combines the simplicity of the traditional linear regression and the flexibility of the non-linear regression. When a certain response variable is estimated, a data subset is taken from the vicinity of a predicted variable, then linear regression or quadratic regression is carried out on the subset, a weighted least square method is adopted during regression, namely the weight of a value closer to an estimation point is larger, finally the value of a corresponding variable is estimated by using an obtained local regression model, and the whole fitting curve is obtained by point-by-point operation by the method. The implementation flow of the Robust local weighted regression algorithm (Robust lose) is as follows:

(1) Selecting proper window number and observing point x _i (i =1,2, \8230;, n), as much as x _i The window width is selected for the center.

(2) Weights are defined for all points within the window, the weights being determined by a weight function.

(3) Using least square method to measure each band weight w _k (x _i ) Observation point (x) of (2) _i ,y _i ) Calculating a regression coefficient alpha (x) _i ) Is estimated by

Then obtain y _i At x _i The fitting value of (b)

(4) Let B (z) be the defined 4-degree power weight function:

let residual of fitted value

S is | e _i Median of | definition

(5) For each i, in (x) _i ,y _i ) Xi used in points _k w _k (x _i ) Replace the original weight w _k (x _i ) Using least squares meansCalculating d-order polynomial fit, calculating new

Generally, when an observation point is fitted by using a Loess algorithm, a polynomial order, a weight function, an iteration number and a window width are very important, wherein the polynomial order, the weight function and the iteration number can be given in advance.

The local weighted regression process and the robustness process of the robust local weighted regression algorithm are respectively realized in an inner loop and an outer loop of the STL algorithm. Fig. 1 shows a flow diagram of an STL algorithm in accordance with an embodiment of the present application. As shown in fig. 1, the inner loop step of the STL algorithm includes:

and S100, assigning an initial value.

Trend item T _v Is given an initial value of 0, i.e.

And S101, removing the trend item.

From the original time series Y _v Subtract the sequence of trend terms in the k-1 iteration

Wherein k =1,2, \8230;, inner, inner is the inner layer cycle number, the original time sequence Y _v Is denoted as N.

And S102, smoothing the periodic subsequence.

The sample points at the same position of each period in the original time sequence form a subsequence (i.e. a periodic subsequence), and it can be calculated that such a periodic subsequence has n _p And (4) respectively. And smoothing each periodic subsequence by using Loess algorithm, wherein the smoothing parameter of Loess is recorded as parameter _c Respectively extending a time point forward and backward, combining the smooth results to obtain a temporary periodic component in the kth iteration

Wherein v = -n _p +1,…,1,2,…N,…,N+n _p Temporary periodic component

Has a length of N +2N _p 。

S103, low-pass filtering of the periodic subsequence.

For the temporary periodic component obtained in step S102

Sequentially making a length of n _p ，n _p 3, then smoothing by using Loess algorithm, and marking the smoothing parameter of Loess as parameter _l Obtaining the low flux of the periodic subsequence in the kth iteration

Wherein v =1,2, \ 8230, N,

is N.

And S104, removing the trend of the smooth cycle subsequence.

From temporary periodic components

The low flux obtained in step S103 is subtracted

Obtaining the periodic item sequence in the k iteration

Namely that

And S105, removing the period item.

From the original time series Y _v The periodic term sequence obtained in step S104 is subtracted

Obtaining intermediate sequences

Namely that

And S106, smoothing the trend.

Using Loess algorithm to process the intermediate sequence obtained in step S105

Performing smoothing, and marking the smoothing parameter of Loess as parameter _t Obtaining the trend item sequence in the kth iteration

S107, whether convergence occurs is judged.

Judging the periodic item sequence in the k iteration

And trend item sequence

Whether the k-th iteration is converged or not, if so, stopping the iteration, and sequencing the periodic items in the k-th iteration

As original time series Y _v Corresponding periodic item sequence S _v The sequence of trend terms in the k-th iteration is

As original time series Y _v Corresponding trend term sequence T _v From the original time series Y _v Minus the corresponding periodic term sequence S _v And a sequence of trend terms T _v The original time series Y can be obtained _v Corresponding residual term sequence R _v I.e. R _v ＝Y _v -S _v -T _v (ii) a If not, the process returns to step S101 to continue the iteration.

The outer loop of the STL algorithm is mainly used to adjust the weights, and if there are outliers in the time series, the residual terms will be larger. Defining the weight function w as:

h＝6*median(|R _v |)

w＝(1-(|R _v |/h)^2)^2

in the inner loop of each iteration, when the stress regression is performed in step S102 and step S106, the neighborhood weight needs to be updated to w, so as to reduce the influence of the abnormal value on the regression.

In order to enable the STL algorithm to have enough robustness, the inner-layer loop and the outer-layer loop are designed, and when the number of the inner-layer loops is large enough, the trend item sequence and the period item sequence are converged at the end of the inner-layer loops; if no outliers are evident in the time series, the outer loop times can be set to 0.

In the STL decomposition algorithm, the Loess algorithm employed in the above steps S102, S103, and S106 performs regression prediction only by the sequence itself and the peripheral values, resulting in inaccurate prediction, thereby reasonably completing the airline commission return protocol.

In order to reasonably complete the airline department commission returning agreement, the embodiment of the application provides a ticket sales order prediction method.

Fig. 2 shows a flow chart of a method for predicting a ticket sales order according to an embodiment of the present application. As shown in fig. 2, the method includes:

s201, obtaining air ticket order history data.

Illustratively, the ticket order history data may include sales volume, total ticket price, flight volume, travel information, amount of returned commissions, etc. of ticket sales orders over a past period of time; for example, data may be included of sales volume, total ticket face price, leg volume, travel information, amount of returned commissions, etc. for a ticket sales order two years ahead from the current time.

S202, obtaining target prediction data according to the historical target data and at least one characteristic; the historical target data comprises first order quantity information of the air ticket sales orders in a first time period in the air ticket order historical data; the at least one characteristic is constructed according to the travel information of the air ticket sales order in the air ticket order historical data; the target forecast data includes second order quantity information for the airline ticket sales orders over a specified time period in at least one airline turnback agreement.

Illustratively, the first order quantity information may include information reflecting the order quantity, such as a total fare and/or a flight quantity, and the historical target data may include a total fare and/or a flight quantity of the ticket sales order in a first time period corresponding to at least one flight repayment agreement in the ticket order historical data, that is, a total fare and/or a flight quantity of the ticket sales order satisfying the designated flight, the designated departure-arrival city, and the designated flight in the flight repayment agreement in the first time period. Illustratively, the first order quantity information may include a commission return amount, and the historical target data may include a commission return amount obtained by the ticket sale order in a first time period corresponding to at least one airline commission return agreement in the ticket order historical data, namely, a commission return amount obtained by the ticket sale order meeting a specified airline department, a specified airline section, a specified flight number and a specified slot combination in the airline commission return agreement in the first time period. Illustratively, the first time period may be a period of time advanced from the current time, for example, 2 years advanced from the current time, such as 6 months and 27 days at 2022, and the first time period may be 27 days at 6 months and 27 days at 2022 at 6 months and 27 days at 2020. Illustratively, the historical target data may be data that is counted by day.

For example, a starting and stopping time period can be specified in the flight return commission protocol, and a target to be completed in the starting and stopping time period, if the current time is not within the starting and stopping time period specified in the flight return commission protocol, the specified time period corresponding to the target prediction data can be the starting and stopping time period specified in the flight return commission protocol; for example, the current time is 2021 year, 12 month, 1 day, the start-stop time specified by the airline department return commission agreement is from 2022 year, 1 month, 1 day to 2022 year, 12 month, 31 day, and the specified time period may be from 2022 year, 1 month, 1 day to 2022 year, 12 month, 31 day; if the current time is within the starting and stopping time period specified in the flight return commission protocol, the specified time period corresponding to the target prediction data can be from the current time to the ending time specified in the flight return commission protocol; for example, the current time is 2022, 8/1/day, the start-stop time specified by the airline company return commission agreement is from 2022, 1/day to 2022, 12/31/day, and the specified time period may be from 2022, 8/1/day to 2022, 12/31/day.

Illustratively, the second order quantity information may include a total fare and/or a leg quantity, and the objective forecast data may include a total fare and/or a leg quantity for a ticket sales order for a specified airline department, a specified departure-arrival city, and a specified leg within a specified time period in at least one airline turnback agreement. Illustratively, the second order quantity information may include a return amount, and the target forecast data may include a return amount available for a ticket sales order for a specified flight, a specified flight segment, a specified flight number, and a specified slot combination for a specified time period in at least one flight return agreement.

For example, in step S202, the at least one feature may be constructed according to the itinerary information of the air ticket sales order corresponding to the first order quantity information in the air ticket order history data, and the itinerary information may include one or more items of date information, holiday information, city information, and business trip application information.

As one example, the features may be constructed from date information and holiday information for the ticket sales order in the ticket order history data. For example, the holiday feature can be constructed based on information such as whether the departure date of the ticket sales order in the ticket order history data is a holiday, whether the departure date is one, two, or three days before and after the holiday. Therefore, the holiday characteristics are constructed according to the date information and the holiday information of the air ticket sales order in the air ticket order historical data, and the holiday characteristics are used for predicting target prediction data, so that a more accurate prediction result is obtained.

As one example, the features may be constructed from city information and business trip application information for a ticket sales order in the ticket order history data. For example, the business trip application number feature may be constructed according to business trip cities and business trip application numbers of the business trip cities corresponding to the ticket sales orders in the historical data of the ticket orders. As an example, the departure cities corresponding to the airline ticket sales orders in the last two years may be sorted according to the order number of each city as the departure place from high to low, and a departure city list may be established; the arrival cities corresponding to the air ticket sale orders in the last two years can be sorted from high to low according to the number of the orders serving as destinations of each city, and an arrival city list is established; sorting the round-trip cities corresponding to the air ticket sale orders in the last two years according to the order quantity of the two cities as departure places-arrival places from high to low, and establishing a departure-arrival city list; the business trip application number feature can be constructed according to at least one of the business trip application number taking the city A as the departure place, the business trip application number taking the city B as the destination, the business trip application number taking the city C as the departure place and the business trip application number taking the city D as the destination and the like; wherein, the city A can be the first 11 cities in the starting city list; city B may be the first 11 cities in the list of reached cities; city C and city D may be departure-arrival cities of the first 15 in the list of departure-arrival cities. Therefore, the airline company commission returning agreement is related to the city and the airline, the business trip application number characteristic is constructed according to the business trip application number of the business trip city and the business trip city corresponding to the air ticket sale order in the air ticket order historical data, and the business trip application number characteristic is used for predicting target prediction data, so that the accuracy of a prediction result is improved.

S203, determining whether to recommend a protocol flight to the user according to the target prediction data; wherein the agreement flight comprises a flight corresponding to the at least one airline commission-returning agreement.

Illustratively, the target prediction data may include a total ticket face price of the air ticket sales order in a specified time period in at least one airline commission return agreement, further, a predicted target completion rate may be calculated from the predicted total ticket face price (target completion rate = total ticket face price of the air ticket sales order in the predicted specified time period/total ticket face price specified for the specified time period in the corresponding airline commission return agreement), and a predicted commission return amount may be calculated from the predicted total ticket face price and the specification of the corresponding airline commission return agreement, so that whether to recommend an agreement flight to the user may be determined from the predicted target completion rate and/or the predicted commission return amount of the airline commission return agreement. Illustratively, if the predicted target completion rate reaches a preset threshold, determining to recommend an agreement flight to the user; otherwise, no flight recommendation is made to the user. Preferably, the preset threshold value interval may be [0.9,0.95]. Illustratively, if the predicted commission return amount reaches the desired commission return amount, determining to recommend an agreement flight to the user; otherwise, not recommending the flight to the user; the desired amount of commission returned can be set according to business requirements.

As an example, the starting and ending time period specified by a certain airline department commission-returning protocol is 1/2022-12/31/2022, and the total ticket price interval specified for obtaining the airline ticket sales order for commission-returning is 1000-9999 ten thousand yuan, namely the total ticket price of the airline ticket sales order meeting the airline department commission-returning protocol specification in the time period of 1/2022-12/31/2022 needs to reach 1000 ten thousand yuan to obtain commission-returning; if the current time is 1/2022, the specified time period may be 1/2022-12/31/2022, and the preset threshold may be 0.92, and if the ratio obtained by dividing the total ticket face price of the air ticket sales order according to the airline department return commission agreement by 1000 ten thousand dollars in the predicted time period from 1/2022 to 12/31/2022 exceeds 0.92, it indicates that the minimum target specified by the airline department return commission agreement is very likely to be realized, and the flight specified in the airline department return commission agreement (i.e., the agreement flight) can be recommended to the user when the user inquires the flight, so as to improve the completion rate of the airline department return commission agreement, and thus increase the profit brought by the department return commission agreement.

Therefore, target prediction data of the air ticket sales order is obtained according to the historical target data and at least one characteristic of the air ticket sales order, and the accuracy of a prediction result is improved; whether the scheduled flight in the airline department commission returning protocol is recommended to the user is determined according to the target prediction data of the air ticket sales order, so that the airline department commission returning protocol can be reasonably completed, the completion rate of the airline department commission returning protocol is improved, and the income brought by the airline department commission returning protocol is improved.

In one possible implementation, under the condition that the agreement flight is determined to be recommended to the user, obtaining a recommendation index of at least one agreement flight according to the user portrait and the agreement flight portrait; the user portrait is constructed according to historical order information of a user; the protocol flight portrait is constructed according to the information of the protocol flight; and recommending agreement flights to the user according to the recommendation index.

It should be noted that, under the condition that the protocol flight is determined to be recommended to the user, if the flight search is performed by the user and no protocol flight exists in the flight list meeting the user search condition, the recommendation process is ended and the flight recommendation is not performed to the user; if there are agreement flights in the flight list, then the following recommendation flow is entered.

Illustratively, the user portrait can be constructed according to flight departure time period, arrival time period, driver, model, price index and other information in the historical order information of the user; wherein, the price index may represent the proximity of the ticket price of the flight at the time of the current user search to the lowest price in the searched result, for example, the price index = (the economy class no discount price of the flight-the ticket price of the flight at the time of the current user search)/(the economy class no discount price of the flight-the lowest price of the ticket searched by the current user).

FIG. 3 illustrates a schematic diagram of a user representation in accordance with an implementation of the present application. As shown in fig. 3, the user profile may include information such as navigation preference, departure time preference, model preference, price index preference, etc., and preference coefficients of various preferences may be calculated by dividing the frequency of the belonged category of the user's historical orders by the total number of orders; for example, the total number of historical orders of a certain user is n, wherein the order number of the A navigation department is k, the order number of the B navigation department is s, and the order number of the C navigation department is n-k-s, so that the preference coefficient of the A navigation department of the user is k/n, the preference coefficient of the B navigation department is s/n, the preference coefficient of the C navigation department is (n-k-s)/n, and the preference coefficients of other navigation departments are all 0. Preference coefficients for departure time preferences, model preferences, price index preferences, etc. may also be calculated in a similar manner. Illustratively, if there is no historical order information of the user, a user portrait can be constructed according to the historical order information of the co-department colleagues of the user, that is, the co-department colleague portrait of the user can be used as the user portrait of the user; if the historical order information of the colleagues in the same department of the user does not exist, a user portrait can be constructed according to the historical order information of the ticket purchased by the company of the user, and the company portrait of the company of the user can be used as the user portrait of the user.

Illustratively, the agreement flight portrait may be constructed according to information about whether the agreement flight is a straight flight, a departure time period, an arrival time period, a flight department, a price index, a model, etc. Illustratively, the preference characteristics of the agreement flight can be constructed according to the preference types contained in the user portrait information, and the preference characteristics of the agreement flight can be in one-to-one correspondence with the preference types of the users; the value of the characteristic value corresponding to each preference feature of the agreement flight is 0 or 1 according to the preference coefficient corresponding to each preference type in the user portrait information and the agreement flight information, wherein the number of the characteristic values corresponding to each preference feature of the agreement flight is the same as the number of the preference coefficients corresponding to each preference type in the user portrait information. For example, the user portrait information contains navigation preference and model preference, wherein the navigation preference comprises three preference coefficients, namely an A navigation preference coefficient, a B navigation preference coefficient and a C navigation preference coefficient, and the model preference comprises three preference coefficients, namely a large preference coefficient, a medium preference coefficient and a small preference coefficient; the navigation department to which the protocol flight belongs is an A navigation department, and the model of the protocol flight is small; the preference characteristics of the protocol flight constructed according to the user portrait information comprise a navigation department preference characteristic and a machine type preference characteristic, the navigation department preference characteristic comprises an A navigation department, a B navigation department and a C navigation department (which respectively correspond to an A navigation department preference coefficient, a B navigation department preference coefficient and a C navigation department preference coefficient in the user portrait information), and the protocol flight belongs to the A navigation department, so that the A navigation department can be taken as 1, and the B navigation department and the C navigation department can be taken as 0; the model preference characteristics comprise large, medium and small (corresponding to large preference coefficient, medium preference coefficient and small preference coefficient in user portrait information respectively), and as the model of the protocol flight is small, the small value can be 1, and the large value and the medium value can be 0. According to the preference coefficient corresponding to each preference type in the user portrait information and the characteristic value corresponding to each preference characteristic of the protocol flight, the recommendation index corresponding to each preference characteristic of the protocol flight can be calculated according to the following formula:

wherein, c _i Recommendation index corresponding to the ith preference feature representing the agreement flight, S _i The number of characteristic values corresponding to the ith preference characteristic representing the flight agreement, a _i,j A j-th preference coefficient, x, corresponding to a preference type of the user (preference type corresponding to the i-th preference feature of the agreement flight) _i,j And j characteristic value corresponding to the ith preference characteristic of the flight in agreement. For example, the 1 st preference feature of the agreement flight is a model preference feature (i.e., i =1 in the above formula), the model of the agreement flight is large, and the model preferences of the user include a large preference coefficient, a medium preference coefficient, and a small preference coefficient, which are 3 preference coefficients, wherein the large preference coefficient is 0.3 (i.e., a in the above formula) _1,1 = 0.3), and the medium preference coefficient is 0.4 (i.e., a in the above formula) _1,2 = 0.4), the mini-preference factor is 0.3 (i.e., a in the above formula) _1,3 = 0.3), the number of eigenvalues corresponding to the model preference characteristics of the agreement flight is 3 (i.e., S in the above formula) ₁ = 3), the medium-large value of the model preference characteristic is 1 (namely x in the formula) _1,1 = 1), medium and small values of 0 (i.e. x in the above formula) _1,2 ＝0，x _1,3 = 0); the model recommendation index c of the protocol flight can be calculated according to the formula ₁ ＝0.3×1+0.4×0+0.3×1＝0.3。

After the recommendation index corresponding to each preference feature of the agreement flight is obtained through calculation, the recommendation indexes corresponding to the preference features can be added to obtain the recommendation index of the agreement flight, and the calculation formula is as follows:

where N represents the number of preference features of the agreement flight (i.e., the number of preference types of the user). After the recommendation index of each protocol flight is calculated, the protocol flight with the highest recommendation index can be recommended to the user; and outputting a protocol flight recommendation list to the user, wherein the protocol flights in the recommendation list can be arranged from high to low according to the recommendation index. For example, when a user searches flights, agreement flights meeting the user search conditions comprise agreement flights a, b and c, the recommendation index of the agreement flight a is calculated to be 2.6, the recommendation index of the agreement flight b is 1.5, and the recommendation index of the agreement flight c is 2.1, the agreement flight a with the highest recommendation index can be recommended to the user, or the agreement flights can be sorted from high to low according to the recommendation indexes, and an agreement flight recommendation list is output to the user according to the sorting of the agreement flights a, the agreement flights c and the agreement flights b.

FIG. 4 illustrates a schematic diagram of a user representation and a protocol flight representation in accordance with an implementation of the present application. FIG. 4 (a) is a schematic diagram of a user profile according to an embodiment of the present application, as shown in FIG. 4 (a), including 4 preference types including driver preference, departure time preference, model preference, and price index preference. Fig. 4 (b) shows a schematic diagram of an agreement flight portrait according to an embodiment of the present application, as shown in fig. 4 (b), the department of the agreement flight is a department, the departure time is 11. According to the preference type contained in the user portrait information and the information of the agreement flight, a navigation preference feature, a departure time preference feature, a model preference feature and a price index preference feature of the agreement flight can be constructed (namely the number of the preference features of the agreement flight in the formula is N = 4). If the agreement flight belongs to the department of navigation A, the department of navigation A takes a value of 1, the department of navigation B and the department of navigation C take a value of 0 in the preference characteristics of the department of navigation, and the department recommendation index C of the agreement flight can be calculated according to the preference coefficient corresponding to the preference of the department of navigation of the user ₁ =0.3 × 1+0.2 × 0+0.5 × 0=0.3. The departure time of the agreement flight is 11, then in the departure time preference feature, the value of the departure time period [8, 00, 12)Index of recommendation c ₂ 0.1 × 0+0.2 × 1+0.3 × 0+0.2 × 0+0.1 × 0=0.2. If the model of the protocol flight is large, the large value is 1 in the model preference characteristics, the medium and small values are 0, and the model recommendation index c of the protocol flight can be calculated according to the preference coefficient corresponding to the model preference of the user ₃ =0.4 × 1+0.3 × 0=0.4. The price index of the agreement flight is 0.5, then in the price index preference feature, the price index interval (0.4, 0.6)]The value is 1, the price index interval (0, 0.2)]、(0.2,0.4]、(0.6,0.8]、(0.8,1]The value is 0, and the price index recommendation index c of the flight in the agreement can be calculated according to the preference coefficient corresponding to the price index preference of the user ₄ =0.2 × 0+0.1 × 1+0.3 × 0+0.2 × 0=0.1. The calculated navigation recommendation index c of the protocol flight ₁ Recommended departure time index c ₂ Model recommendation index c ₃ Price index recommendation index c ₄ Adding to obtain the recommendation index C = C of the flight in the agreement ₁ +c ₂ +c ₃ +c ₄ ＝0.3+0.2+0.4+0.1＝1。

Therefore, under the condition that the agreement flight is determined to be recommended to the user, the recommendation index of the agreement flight is calculated according to the user portrait and the agreement flight portrait, and the agreement flight is recommended to the user according to the recommendation index, so that the ordering rate of the user for the agreement flight ticket is improved, the momentum of the airline company commission returning protocol which does not finish the target can be realized, the completion rate of the airline company commission returning protocol is improved, and the income brought by the airline company commission returning protocol is improved.

In a possible implementation manner, the target prediction data in step S202 further includes third order quantity information of the air ticket sales order in a second time period corresponding to the at least one airline commission-return agreement; and the third order quantity information of the air ticket sales order in the second time period corresponding to the at least one airline turnback agreement is used for evaluating the airline turnback agreement in the second time period.

Illustratively, the second time period may be a time period after the expiration of the airline commission return protocol. Illustratively, the third order quantity information may include a total fare and/or a leg quantity, and the target forecast data may include a total fare and/or a leg quantity for a ticket sales order for a designated flight, a designated departure-arrival city, and a designated leg corresponding to the at least one flight return commission agreement.

As an example, the start-stop time specified by a certain airline department return commission agreement is 1/2022-12/31/2022, the second time period may be 1/2023-12/31/2023, the target forecast data is the total ticket face price and/or the flight segment amount of the air ticket sales order meeting the airline department return commission agreement specification in the time period of 1/2023-12/31/2023, and the target forecast data can be output to the business personnel, so as to provide reference for the airline department return commission agreement in the time period of 1/2023-12/31/2023, thereby enabling the business personnel to reasonably establish the airline department return commission agreement in the next stage.

Therefore, by predicting the order quantity information of the airline department commission returning protocol in a period of time in the future, reference can be provided for the airline department commission returning protocol in the next stage, and the establishment of a more reasonable airline department commission returning protocol is facilitated, so that the completion rate of the airline department commission returning protocol is improved, and the income brought by the airline department commission returning protocol is improved.

An exemplary implementation of obtaining the target prediction data according to the historical target data and the at least one feature in step S202 is described below.

Fig. 5 is a flowchart illustrating a method for predicting an air ticket sales order according to an embodiment of the present application, and as shown in fig. 5, in step S102, obtaining target prediction data according to historical target data and at least one feature may include the following steps:

s2021, obtaining a historical time sequence according to the historical target data.

Illustratively, historical target data counted by day (or counted by other time units) can be arranged according to time sequence to obtain a historical time sequence. Table 1 shows a set of historical target data according to an embodiment of the present application. As shown in table 1, the current time is 2022, 6 months and 27 days, and the historical target data is the total ticket face price and the return commission amount of the air ticket sales order corresponding to the same airline department return commission agreement, which is pushed 2 years ahead from the current time, namely the total ticket face price of the air ticket sales order meeting the airline department return commission agreement, the departure-arrival city and the designated flight segment in the 6 months and 27 days in 2020 and 2022, and the return commission amount obtained by the air ticket sales order meeting the airline department return commission agreement, the designated flight number and the designated slot combination in the 6 months and 27 days in 2020 and 27 months. For example, for a missing value in the historical target data, the missing value may be replaced by historical data at the same time in a cycle on the missing value based on a method of historical value interpolation; for example, if the total fare for 8 month 1 day in 2021 is missing, the total fare for 8 month 1 day in 2019 may be used instead. The historical target data can be counted by day and arranged by date, and a historical time sequence can be obtained.

TABLE 1

S2022, inputting the historical time sequence and the at least one feature into a preset model to obtain the target prediction data; the preset model is obtained by training based on an STL time sequence decomposition algorithm, so that the STL algorithm is improved.

In a possible implementation manner, the preset model includes a first submodel, a second submodel, and a third submodel; the first sub-model, the second sub-model and the third sub-model are obtained by training based on a regression algorithm; the inputting the historical time series and the at least one characteristic into a preset model to obtain the target prediction data may include:

(1) And obtaining an input sequence according to the historical time sequence and the at least one characteristic.

For example, the feature values corresponding to at least one feature may be arranged in a chronological order, and together with the historical time series, the input series is formed.

For example, for an input sequence, the input data at the same position in each cycle may be grouped into a sub-sequence (i.e., a periodic sub-sequence). Table 2 shows a set of input data according to an embodiment of the present application. As shown in table 2, the input data is the total ticket face price of the air ticket sales order meeting the rules of the airline department return commission agreement and the feature value corresponding to n features in the period from 27 th 6/month in 2020 to 27 th 6/month in 2022, wherein n is larger than or equal to 1, the feature 1 can be the holiday feature in step S202 in fig. 2, for example, and the feature 2 can be the business trip application number feature in step S202 in fig. 2, for example; setting the number of days in a period as 7 days, wherein the period value of the current time can be 1, namely the period value of 2022.06.27 is 1; for the previous period of the current time, namely 2022.06.20-2022.06.26, the period values are sequentially taken as 1,2, \ 8230and 7 in time sequence, namely the period value of 2022.06.20 is taken as 1 and the period value of 2022.06.26 is taken as 7; according to the method, a period value corresponding to each date of 2020.06.27-2022.06.27 can be obtained; all data corresponding to dates having a period value of i (i.e. the same position i) can form a period subsequence c _i Wherein, i =1,2, \82307; for example, 2020.06.29, 2020.07.06, \8230, 2022.06.20 and 2022.06.27, which all have the period value of 1, can constitute the period subsequence c ₁ (ii) a The input data may be arranged in chronological order to obtain an input sequence.

TABLE 2

(2) Acquiring a trend item sequence in the (k-1) th iteration, and acquiring a first intermediate sequence according to the input sequence and the trend item sequence in the (k-1) th iteration; wherein the initial value of k is 1, and the initial value of the trend item sequence is 0.

Illustratively, the sequence of trend terms in the k-1 iteration may be subtracted from the input sequence to obtain a first intermediate sequence.

(3) And inputting the first intermediate sequence into the first submodel to obtain a second intermediate sequence.

Exemplarily, the data at the same position in each period of the first intermediate sequence may be grouped into a period sub-sequence, so as to obtain at least one period sub-sequence corresponding to the first intermediate sequence; each period subsequence can be input into the first sub-model for regression, and each period subsequence is extended forward and backward by one period, and the results of regression of each period subsequence are combined to obtain a second intermediate sequence.

As an example, taking the input data in Table 2 as an example, the total fare corresponding to the dates with the period values of 1 in 2020.06.29, 2020.07.06, \ 8230, 2022.06.20 and 2022.06.27 and the feature values corresponding to the features can form a period subsequence c ₁ C is mixing ₁ The first sub-model is input to perform regression, and the prediction data obtained after the first sub-model is extended forward and backward for one cycle respectively comprises prediction data of 2020.06.22 (extended forward for one cycle) and prediction data of 2022.07.04 (extended backward for one cycle).

Illustratively, the first sub-model is a regression model obtained by training; in the training process, a part of data in the input sequence can be used as a training set, and the rest part of data can be used as a test set; for example, the first 80% of the data may be used as a training set, and the second 20% of the data may be used as a testing set; at least one periodic subsequence corresponding to the training set can be used as a training sample to train the preset regression model, and therefore the first sub model is obtained.

(4) And obtaining a third intermediate sequence according to the second intermediate sequence.

Illustratively, the second intermediate sequence may be low-pass filtered moving average sequentially, e.g. sequentially of length n _p ，n _p 3 to obtain a third intermediate sequenceWherein n is _p The number of the periodic subsequences corresponding to the second intermediate sequence. Sequentially carrying out the second intermediate sequence with the length of n _p ，n _p The calculation procedure of the moving average of 3 is as follows:

wherein N represents the length of the periodic subsequence corresponding to the second intermediate sequence, C represents the second intermediate sequence, and ma1, ma2 and ma3 are respectively the length N of the second intermediate sequence in sequence _p ，n _p And 3, the sequence obtained after the moving average of 3 is the third intermediate sequence, namely the sequence ma 3.

(5) And inputting the third intermediate sequence into the second submodel to obtain a fourth intermediate sequence.

For example, the third intermediate sequence may be input into the second submodel for regression to obtain a fourth intermediate sequence.

Illustratively, the second sub-model is a regression model obtained through training, and in the training process, a part of data in the third intermediate sequence may be used as a training set, and the remaining part of data may be used as a test set, and a preset regression model is trained, so as to obtain the second sub-model; for example, the first 80% of the data may be used as a training set and the second 20% as a testing set.

(6) And obtaining a periodic item sequence and a fifth intermediate sequence in the kth iteration according to the input sequence, the second intermediate sequence and the fourth intermediate sequence.

For example, the fourth intermediate sequence may be subtracted from the second intermediate sequence to obtain a periodic term sequence in the k-th iteration; the periodic term sequence in the kth iteration may be subtracted from the input sequence to obtain the fifth intermediate sequence.

(7) And inputting the fifth intermediate sequence into the third submodel to obtain a trend item sequence in the kth iteration.

Illustratively, the fifth intermediate sequence may be input into the third sub-model for regression, resulting in a sequence of trend terms in the kth iteration.

Illustratively, the third sub-model is a regression model obtained through training, and in the training process, a part of data in the fifth intermediate sequence may be used as a training set, and the remaining part of data may be used as a test set, and a preset regression model is trained, so as to obtain a third sub-model; for example, the first 80% of the data may be used as a training set, and the second 20% of the data may be used as a testing set.

(8) Judging whether the periodic item sequence in the kth iteration and the trend item sequence in the kth iteration are converged, stopping the iteration under the condition that the periodic item sequence in the kth iteration and the trend item sequence in the kth iteration are converged, determining the periodic item sequence in the latest iteration as the periodic item sequence corresponding to the input sequence, and determining the trend item sequence in the latest iteration as the trend item sequence corresponding to the input sequence; otherwise, adding 1 to the value of k, and returning to execute the step of acquiring the trend item sequence in the (k-1) th iteration.

For example, the input sequence may be subtracted by the periodic term sequence in the k-th iteration and then subtracted by the trend term sequence in the k-th iteration to obtain the residual term sequence in the k-th iteration. For example, it may also be determined whether to stop the iteration according to whether the sequence of residual terms in the k-th iteration converges.

(9) And obtaining the target prediction data according to the period item sequence corresponding to the input sequence and the trend item sequence corresponding to the input sequence.

For example, the time sequence decomposition of the input sequence may be completed by subtracting the periodic term sequence corresponding to the input sequence from the input sequence and then subtracting the trend term sequence corresponding to the input sequence to obtain a residual term sequence corresponding to the input sequence; furthermore, the second order quantity information of the air ticket sales order in the specified time period in the airline department return commission protocol can be predicted according to the time series decomposition result of the input sequence, that is, the target prediction data can be obtained through prediction, for example, the second order quantity information of the air ticket sales order in the specified time period in the airline department return commission protocol can be predicted by using the cycle item sequence, the trend item sequence and the residual error item sequence.

Illustratively, the first sub-model, the second sub-model and the third sub-model may be XGBoost models, which are an integrated machine learning model, may be used for regression problems, and have very strong robustness and generalization of model prediction results without feature selection. Illustratively, in the model training process, for each iteration, the first sub-model, the second sub-model, and the third sub-model in the above steps (3), (5), (7) may share a set of hyperparameter values, and the optimal hyperparameter value is based on the minimum Root Mean Square Error (RMSE) on the test set. Illustratively, the hyper-parameters of the XGBoost model may include: max _ depth: the maximum tree depth, which is mainly used to avoid overfitting, when the value is larger, the model can learn more specific and more local samples; min _ child _ weight: the minimum leaf node sample weight sum is also used for avoiding overfitting, and when the value is larger, the model can be prevented from learning local special samples; subsample: controlling the proportion of random sampling of each tree; gamma: the minimum penalty function degradation value required for node splitting may be specified.

In the model training process, the first submodel, the second submodel and the third submodel output in each iteration process may be stored in a prediction file, and after the model training is finished, all the first submodels, the second submodels and the third submodels stored in the prediction file may be compared; selecting the first sub-model with the minimum RMSE on the test set as the trained first sub-model for all the first sub-models stored in the prediction file; selecting the second submodel with the minimum RMSE on the test set as a trained second submodel for all second submodels stored in the prediction file; selecting the third submodel with the minimum RMSE on the test set as a trained third submodel for all the third submodels stored in the prediction file; and obtaining a trained preset model according to the trained first sub-model, the trained second sub-model and the trained third sub-model, so that target prediction data can be predicted through the preset model.

Therefore, the traditional method for directly inputting the historical time sequence into the traditional STL algorithm model for prediction is different from the traditional method for directly inputting the historical time sequence into the traditional STL algorithm model, and the traditional STL algorithm only depends on a single time sequence and is difficult to extract trend items and periodic items with better effects; in the embodiment of the application, the historical time sequence and the characteristics are input into the preset model together to obtain target prediction data, and the extraction method of the periodic item sequence and the trend item sequence in the STL algorithm can be optimized, so that the improvement of the STL algorithm is realized; in addition, the method is different from the existing method that long-period data are not considered, and a prediction model is built only by depending on local data to be predicted; the embodiment of the application solves the problem of dependence on local data, and can obtain a more accurate time series decomposition result, so that the accuracy of the prediction result can be improved, and more accurate target prediction data can be obtained.

FIG. 6 shows a flow chart for building a predictive model according to an embodiment of the present application. As shown in fig. 6, a historical time series may be obtained according to the historical target data, and the process of constructing the historical time series may refer to step S2021 in fig. 5; the feature engineering may include two parts, namely missing value processing and feature construction, and the missing value processing may refer to step S2021 in fig. 5; holiday characteristics and business trip application number characteristics can be constructed according to the travel information of the air ticket sales order in the air ticket order historical data, and the characteristic construction process can refer to the step 202 in the figure 2; since most employees usually submit business trip applications within 2 weeks before business trip, the business trip application number features can be loaded into the construction task of the preset model only within 2 weeks before the driver return commission agreement is finished; the model construction can comprise data set division and algorithm type selection, 80% of data in input data can be selected as a training set for model training, and the rest 20% of data can be selected as a test set for model verification; the algorithm can select an STL algorithm, a historical time sequence and at least one characteristic can be input into an STL algorithm model for training, and the training process can refer to the steps (1) - (9); in the training process of the preset model, a first sub-model, a second sub-model and a third sub-model which are output in each iteration process of the STL algorithm can be stored in a prediction file, and when a period item sequence and a trend item sequence which are output by the STL algorithm model are converged, the training is stopped; after the preset model training is finished, the first sub-model, the second sub-model and the third sub-model stored in the prediction file can be compared; selecting the first sub-model with the minimum RMSE on the test set as the trained first sub-model for all the first sub-models stored in the prediction file; selecting the second submodel with the minimum RMSE on the test set as a trained second submodel for all second submodels stored in the prediction file; selecting a third submodel with the minimum RMSE on the test set as a trained third submodel for all third submodels stored in the prediction file; and obtaining a trained preset model according to the trained first sub-model, the trained second sub-model and the trained third sub-model, so that target prediction data can be predicted through the preset model.

Compared with the existing prediction method based on the STL algorithm and only needing to input the sequence, the sequence local data and the weight, the prediction method provided by the embodiment of the application is improved into a prediction method of total regression, and besides the input sequence, external characteristics such as holiday characteristics and business trip application number characteristics can be input, so that the input of a trend item sequence and a periodic item sequence in a time sequence is formed by the input sequence and the external characteristics, the extraction method of the periodic item sequence and the trend item sequence is optimized, the problem of dependence on the local data is solved, and more accurate time sequence decomposition is realized by a method of accurate decomposition and divide-and-conquer of sequence information, so that the accuracy of a prediction result can be improved, and more accurate target prediction data can be obtained.

The air ticket sales order prediction method provided by the embodiment of the application can be applied to an air ticket sales order prediction system. FIG. 7 illustrates a schematic diagram of an airline ticket sales order prediction system according to an embodiment of the present application. As shown in fig. 7, for a flight department commission-returning agreement, the system predicts the total ticket price of the air ticket sale order in the specified time period according to the total ticket price and/or the flight segment amount (i.e. agreement measurement and settlement data) of the air ticket sale order meeting the regulations of the flight department commission-returning agreement in the historical target data, further calculates the target completion rate (i.e. predicted target completion rate) of each flight department commission-returning agreement at the end of the agreement period, predicts the commission-returning amount of each flight department commission-returning agreement according to the commission-returning data (i.e. agreement award settlement data) of the air ticket sale order meeting the regulations of the flight department commission-returning agreement in the historical target data, and the prediction process can refer to steps S2021-S2022 in fig. 5; the system outputs the protocol recommendation degree of each navigation department commission returning protocol according to the predicted target completion rate of each navigation department commission returning protocol, and the higher the predicted target completion rate of the navigation department commission returning protocol is, the higher the corresponding protocol recommendation degree is; the impulse switch can be set for each airline department commission returning protocol, and business personnel can comprehensively evaluate the current progress and the predicted target completion rate of each airline department commission returning protocol and determine whether the impulse switch of each airline department commission returning protocol needs to be opened or not; for example, if the predicted target completion rate of a flight department commission refund agreement is greater than 90%, the impulse switch of the flight department commission refund agreement can be opened to carry out impulse on the flight department commission refund agreement; illustratively, the momentum switches of multiple airline commission-back protocols can be opened simultaneously; under the condition that the impulse switch of a certain airline department commission returning protocol is determined to be opened, if a user logs in a foreground to inquire an air ticket and there is an agreement flight in a flight list meeting the search condition of the user, an agreement flight portrait can be constructed according to the information of the agreement flight, a user portrait can be constructed according to the historical order information of the user, and the method for constructing the agreement flight portrait and the user portrait can refer to the description corresponding to the figure 3 and the figure 4; calculating recommendation indexes of the protocol flights by combining the protocol flight portrait and the user portrait, wherein the calculation method of the recommendation indexes can refer to the description corresponding to the figure 4; recommending the agreement flight with the highest recommendation index to the user, or arranging according to the recommendation index of the agreement flight from high to low, outputting a recommendation list to the user flight, and recommending the user to place an order through a front page so as to finish the agreement target; if the user subscribes to an agreement flight, a pulse is completed for the flight back commission agreement; this process is called revenue recommendation. It should be noted that, in the case of determining to recommend an agreement flight to the user, if there is no agreement flight in the flight list, the recommendation process is ended, and no flight recommendation is made to the user. The system predicts the total ticket face price interval of the air ticket sales order within a period of time in the future of the airline department commission return agreement according to the agreement metering settlement data, outputs the prediction result to the business personnel, and provides reference for reasonably evaluating the target of the next stage, and the process can be called revenue counseling. The air ticket sales order prediction system in the embodiment of the application can embody the air ticket order data of each airline department commission returning protocol in the air ticket income management system, and the income recommendation and income recommendation are completed by using the air ticket order historical data, so that a set of complete solution is provided for business personnel to solve the target impulse and target formulation of the airline department commission returning protocol.

Fig. 8 (a) - (c) show schematic diagrams of a flight ticket sales order prediction system according to an embodiment of the present application. As shown in fig. 8 (a), the air ticket sales order prediction system in the embodiment of the present application is divided into two parts, i.e., revenue recommendation and revenue advisory, and includes a protocol prediction module, an impulse switch, and a recommendation module. The protocol prediction module can construct a time series prediction model according to historical target data of each airline department commission-returning protocol, predict order quantity information of the air ticket sales order in a specified time period, and the prediction process can refer to the steps S2021-S2022 in the above-mentioned fig. 5; and calculating an agreement recommendation index for each airline department return commission agreement. The impulse switch is controlled by a service person, and the service person can decide whether to carry out impulse on the current agreement according to the order quantity information and/or other prediction data obtained based on the order quantity information, for example, can decide whether to start the impulse switch of a certain airline department commission returning agreement according to the current completion rate, the prediction target completion rate, the prediction commission returning amount and the like of the airline department commission returning agreement; if the system is started, entering a recommending module, otherwise, not entering the recommending module; in addition, order quantity information and/or other forecast data can also provide scientific advice guidance for business personnel to evaluate the total nominal interval for a specified time period. After a momentum switch of a flight driver commission returning protocol is started, when a user inquires a flight of a certain airline, if the flight has an agreement flight, calculating a recommendation index of the agreement flight according to a user portrait formed by historical order information of the user and an agreement flight portrait formed by information of the agreement flight, wherein the method for constructing the agreement flight portrait, the user portrait and the protocol flight recommendation index can refer to the description corresponding to the figure 3 and the figure 4; the flight with the highest recommendation index can be recommended to the user; if the user subscribes to the agreement flight, one agreement impulse is completed. Fig. 8 (b) shows a schematic diagram of a protocol prediction module of the air ticket sales order prediction system according to an embodiment of the present application. As shown in fig. 8 (b), the protocol prediction module trains the prediction model according to the historical target data in the air ticket order historical data; the prediction model may be trained based on an STL algorithm with external features introduced, and the training process may refer to step S2022 in fig. 5; after the trained prediction model is obtained, the protocol prediction module can predict the total ticket price and the commission return amount of the air ticket sales order of each airline department commission return protocol in a specified time period, and can calculate the predicted target completion rate of each airline department commission return protocol according to the predicted total ticket price, so that the protocol recommendation index of each airline department commission return protocol is obtained. Fig. 8 (c) shows a schematic diagram of a recommendation module of the air ticket sales order prediction system according to an embodiment of the present application. As shown in fig. 8 (c), in the case that it is determined that the impulse switch of a certain airline commission returning agreement is turned on, when the user searches for an air ticket, if there is an agreement flight corresponding to the airline commission returning agreement in the flight list meeting the user search condition, the recommending module may construct an agreement flight portrait according to the information of the agreement flight, construct a user portrait according to the historical order information of the user, and construct an agreement flight portrait and a user portrait with reference to the description corresponding to fig. 3 and fig. 4; calculating recommendation indexes of the protocol flights by combining the protocol flight portrait and the user portrait so as to recommend the flights to the user, wherein the calculation method of the recommendation indexes of the protocol flights can refer to the description corresponding to the figure 4; if the flight list meeting the user search condition does not have the agreement flight, the recommendation process is ended, and flight recommendation is not carried out. The air ticket sale order prediction system in the embodiment of the application can predict the target completion rate and the commission return amount of each airline department commission return protocol at the end of the agreement period, business personnel can evaluate whether to carry out impulse on the airline department commission return protocol of the unfinished target according to the prediction result, and the recommendation index of the agreement flight is calculated by combining the agreement flight portrait and the user portrait, so that the agreement flight is recommended to the user to complete the agreement target; the total ticket face price of the protocol in a period of time in the future can be predicted according to the historical data of the air ticket sales orders meeting the rules of the airline department commission-returning protocol, and the prediction result is output to business personnel to provide reference for reasonably evaluating the target of the next stage; meanwhile, the air ticket sales order prediction system in the embodiment of the application trains the prediction model based on the STL algorithm with the introduced external features, namely trains the prediction model through the improved STL algorithm, so that the prediction accuracy of the model can be improved.

The performance of the air ticket sales order prediction method and the air ticket sales order prediction system provided by the application are described below by taking the prediction of the total ticket face price and the impulse of the airline company commission return protocol as examples.

Fig. 9 is a schematic diagram illustrating an air ticket sales order prediction method according to an embodiment of the present application and a prediction curve for predicting a total fare of an airline commission return agreement according to an STL algorithm. Fig. 9 (a) shows a prediction curve diagram for predicting the total fare of the airline commission refund agreement according to the airline ticket sales order prediction method according to the embodiment of the present application, and fig. 9 (b) shows a prediction curve diagram for predicting the total fare of the airline commission refund agreement according to the STL algorithm. Table 3 shows a method for predicting an airline ticket sales order according to an embodiment of the present application and RMSE and summary data error (gap) for predicting a total fare for a airline commission return according to the STL algorithm. As can be seen from fig. 9 and table 3, the prediction method for the air ticket sales order provided in the embodiment of the present application has a higher prediction accuracy and a smaller prediction error compared with the conventional STL algorithm.

TABLE 3

	RMSE	gap
			Method for predicting air ticket sales order	80856	0.057
STL algorithm	176763	0.095

As an example, the agreement period of a return commission agreement of a navigation department is 2021-08-01 to 2021-08-31, and the total nominal price interval agreed with the navigation department is 1000 ten thousand yuan to 9999 ten thousand yuan; that is, in the agreement period, the TMC company can complete the target only if it can complete 1000 ten thousand yuan of total ticket price. At the end of the agreement period, the actual total fare of the air ticket sales order is 1249 ten thousand yuan, and the calculated completion rate is 124.9%. Table 4 shows the total fare prediction by the flight ticket sales order prediction method according to an embodiment of the present application, which predicts the total fare for the first day (i.e. 2021-08-01) of the flight department commission return agreement and the end of the agreement period (i.e. 2021-08-31).

TABLE 4

Date	Predicting total fare (Yuan)	True total nominal price (Yuan)	Whether it meets the standard (minimum target)	Error of the measurement
					2021.08.01	16834470	12490452	Is that	0.347787
2021.08.02	16917320	12490452	Is that	0.35442
					2021.08.03	26767780	12490452	Is that	1.14306
2021.08.04	25984690	12490452	Is that	1.080364
					2021.08.05	25320630	12490452	Is that	1.027199
2021.08.06	24613880	12490452	Is that	0.970616
					2021.08.07	24050160	12490452	Is that	0.925484
2021.08.08	16163380	12490452	Is that	0.294059
					2021.08.09	18723150	12490452	Is that	0.498997
2021.08.10	15911930	12490452	Is that	0.273927
					2021.08.11	15659310	12490452	Is that	0.253702
2021.08.12	14061100	12490452	Is that	0.125748
					2021.08.13	14167490	12490452	Is that	0.134266
2021.08.14	13653070	12490452	Is that	0.093081
					2021.08.15	13737790	12490452	Is that	0.099863
2021.08.16	13626490	12490452	Is that	0.090953
					2021.08.17	13484090	12490452	Is that	0.079552
2021.08.18	13323810	12490452	Is that	0.066717
					2021.08.19	12892080	12490452	Is that	0.032155
2021.08.20	12862440	12490452	Is that	0.029782
					2021.08.21	12783910	12490452	Is that	0.023495
2021.08.22	12635940	12490452	Is that	0.011648
					2021.08.23	12508690	12490452	Is that	0.00146
2021.08.24	11517810	12490452	Is that	0.077871
					2021.08.25	11489770	12490452	Is that	0.080116
2021.08.26	11303190	12490452	Is that	0.095053
					2021.08.27	12203360	12490452	Is that	0.022985
2021.08.28	11885860	12490452	Is that	0.048404
					2021.08.29	12257470	12490452	Is that	0.018653
2021.08.30	12420140	12490452	Is that	0.005629
					2021.08.31	12538770	12490452	Is that	0.003868
Average	12557154	12490452	Is that	0.046365

From the predicted data in table 4, it can be seen that the first day from the beginning of the protocol, i.e. the prediction, can achieve 1000 ten thousand yuan of goal, and the average error between the predicted value and the real value is only 4%. According to the method for predicting the air ticket sales order, the external characteristics are introduced, the historical time sequence and the characteristics are input into the preset model together to obtain the target prediction data, the method for extracting the periodic item sequence and the trend item sequence in the STL algorithm can be optimized, the problem of dependence on local data is solved, and the accuracy of the prediction result is improved.

Fig. 10 is a schematic diagram illustrating target lift rates after three airline department commission return protocols are pulsed by the airline ticket sales order prediction system according to an embodiment of the present application. As shown in fig. 10, after the air ticket sales order prediction system provided by the embodiment of the present application is online on 12/1/2021, by 31/3/2022, the platform carries out momentum on three airline department commission return agreements through the revenue recommendation subsystem of the air ticket sales order prediction system, and by agreement flight recommendation, the recommendation ordering target improvement rates of 1.00%, 1.12%, and 0.76% are respectively realized. The target impulse for a 346 ten thousand dollar ticket sales order is accumulated. The air ticket sale order prediction method provided by the embodiment of the application can provide air ticket income management for a TMC company, and business personnel can determine whether to carry out impulse on the unfinished target airline company commission returning protocol according to the prediction data of the air ticket sale order prediction system, so that the completion rate of the airline company commission returning protocol is improved, and the income brought by the airline company commission returning protocol is improved.

Based on the same inventive concept of the above method embodiment, the embodiment of the present application further provides an air ticket sales order prediction apparatus, which can be used to execute the technical solution described in the above method embodiment. For example, the steps of the methods shown in fig. 2, 5 or 6 described above may be performed.

Fig. 11 is a block diagram illustrating an arrangement for predicting an air ticket sales order according to an embodiment of the present application, and as shown in fig. 11, the arrangement includes: an obtaining module 1101, configured to obtain air ticket order history data; a prediction module 1102, configured to obtain target prediction data according to the historical target data and the at least one feature; the historical target data comprises first order quantity information of the air ticket sales orders in a first time period in the air ticket order historical data; the at least one characteristic is constructed according to the travel information of the air ticket sales order in the air ticket order historical data; the target forecast data comprises second order quantity information of the air ticket sales orders in a specified time period in at least one airline department commission-returning protocol; a recommending module 1103, configured to determine whether to recommend a protocol flight to the user according to the target prediction data; wherein the agreement flight comprises a flight corresponding to the at least one airline commission-returning agreement.

In the embodiment of the application, target prediction data of the air ticket sales order is obtained according to historical target data and at least one characteristic of the air ticket sales order, and the at least one characteristic is constructed according to the travel information of the air ticket sales order in the historical data of the air ticket order, so that the accuracy of a prediction result is improved; whether the scheduled flight in the airline department commission returning protocol is recommended to the user is determined according to the target prediction data of the air ticket sales order, so that the airline department commission returning protocol can be reasonably completed, the completion rate of the airline department commission returning protocol is improved, and the income brought by the airline department commission returning protocol is improved.

In a possible implementation manner, the recommending module 1103 is further configured to: obtaining a recommendation index of at least one agreement flight according to the user portrait and the agreement flight portrait under the condition of determining that the agreement flight is recommended to the user; the user portrait is constructed according to historical order information of a user; the protocol flight portrait is constructed according to the information of the protocol flight; and recommending agreement flights to the user according to the recommendation index.

In a possible implementation manner, the prediction module 1102 is further configured to: obtaining a historical time sequence according to the historical target data; inputting the historical time sequence and the at least one characteristic into a preset model to obtain the target prediction data; and the preset model is obtained by training based on an STL time series decomposition algorithm.

In a possible implementation manner, the preset model includes a first submodel, a second submodel, and a third submodel; the first sub-model, the second sub-model and the third sub-model are obtained by training based on a regression algorithm; the prediction module is further configured to: obtaining an input sequence according to the historical time sequence and the at least one characteristic; acquiring a trend item sequence in the (k-1) th iteration, and acquiring a first intermediate sequence according to the input sequence and the trend item sequence in the (k-1) th iteration; wherein the initial value of k is 1, and the initial value of the trend item sequence is 0; inputting the first intermediate sequence into the first submodel to obtain a second intermediate sequence; obtaining a third intermediate sequence according to the second intermediate sequence; inputting the third intermediate sequence into the second submodel to obtain a fourth intermediate sequence; obtaining a periodic item sequence and a fifth intermediate sequence in the kth iteration according to the input sequence, the second intermediate sequence and the fourth intermediate sequence; inputting the fifth intermediate sequence into the third submodel to obtain a trend term sequence in the kth iteration; judging whether the periodic item sequence in the kth iteration and the trend item sequence in the kth iteration are converged, stopping the iteration under the condition that the periodic item sequence in the kth iteration and the trend item sequence in the kth iteration are converged, determining the periodic item sequence in the latest iteration as the periodic item sequence corresponding to the input sequence, and determining the trend item sequence in the latest iteration as the trend item sequence corresponding to the input sequence; otherwise, adding 1 to the value of k, and returning to execute the step of acquiring the trend item sequence in the (k-1) th iteration; and obtaining the target prediction data according to the period item sequence corresponding to the input sequence and the trend item sequence corresponding to the input sequence.

In a possible implementation manner, the prediction module 1102 is further configured to: subtracting the fourth intermediate sequence from the second intermediate sequence to obtain a periodic term sequence in the kth iteration; and subtracting the periodic term sequence in the kth iteration from the input sequence to obtain a fifth intermediate sequence.

In one possible implementation, the objective prediction data further includes third order quantity information for the airline ticket sales order for a second time period corresponding to the at least one airline turnback agreement; and the third order quantity information of the air ticket sales order in the second time period corresponding to the at least one airline turnback agreement is used for evaluating the airline turnback agreement in the second time period.

In one possible implementation, the first order quantity information and the second order quantity information include one or more of a total fare, a leg quantity, and a commission return amount.

In one possible implementation, the travel information includes one or more of date information, holiday information, city information, business trip application information.

For technical effects and specific descriptions of the air ticket sales order prediction apparatus shown in fig. 11 and various possible implementations thereof, reference may be made to the air ticket sales order prediction method described above, and details are not described here.

It should be understood that the division of the modules in the above air ticket sales order prediction apparatus is only a division of logical functions, and the actual implementation may be wholly or partially integrated into one physical entity, or may be physically separated. In addition, the modules in the device can be realized in the form of calling software by a processor; for example, the apparatus includes a processor, the processor is connected to a memory, the memory stores instructions, the processor calls the instructions stored in the memory to implement any one of the above methods or implement the functions of the modules of the apparatus, wherein the processor is a general-purpose processor such as a Central Processing Unit (CPU) or a microprocessor, and the memory is an internal memory of the apparatus or an external memory of the apparatus. Alternatively, a module in the apparatus may be implemented in a form of a hardware circuit, and a part or all of the functions of the module may be implemented by designing the hardware circuit, which may be understood as one or more processors; for example, in one implementation, the hardware circuit is an application-specific integrated circuit (ASIC), and the functions of some or all of the above modules are implemented through the design of the logical relationship of elements in the circuit; for another example, in another implementation, the hardware circuit may be implemented by a Programmable Logic Device (PLD), and may include a large number of logic gates, for example, a Field Programmable Gate Array (FPGA), and the connection relationship between the logic gates is configured by a configuration file, so as to implement the functions of some or all of the above modules. All the modules of the device can be realized in a form of calling software by a processor, or in a form of calling software by a hardware circuit, or in a form of calling software by a processor, and the rest is realized in a form of calling hardware circuit.

In the embodiment of the present application, the processor is a circuit having a signal processing capability, and in one implementation, the processor may be a circuit having an instruction reading and executing capability, such as a CPU, a microprocessor, a Graphics Processing Unit (GPU), a Digital Signal Processor (DSP), a neural-Network Processing Unit (NPU), a Tensor Processing Unit (TPU), or the like; in another implementation, the processor may implement certain functions through the logical relationship of hardware circuits, which may be fixed or reconfigurable, such as a hardware circuit implemented by an ASIC or PLD, such as an FPGA. In the reconfigurable hardware circuit, the process of loading the configuration document by the processor to implement the configuration of the hardware circuit may be understood as a process of loading instructions by the processor to implement the functions of some or all of the above modules.

It is seen that the modules in the above apparatus may be one or more processors (or processing circuits) configured to implement the above embodiment methods, for example: CPU, GPU, NPU, TPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms. In addition, all or part of the modules in the above apparatus may be integrated together, or may be implemented independently, which is not limited in this respect.

An embodiment of the present application further provides an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to implement the method of the above embodiment when executing the instructions. Illustratively, the steps of the methods illustrated in fig. 2, 5 or 6 described above may be performed.

Fig. 12 is a schematic structural diagram of an electronic device according to an embodiment of the present application, and as shown in fig. 12, the electronic device may include: at least one processor 1701, communication lines 1702, memory 1703, and at least one communication interface 704.

The processor 1701 may be a general purpose central processing unit, microprocessor, application specific integrated circuit, or one or more integrated circuits configured to control the execution of programs in accordance with the present invention; the processor 1701 may also include a heterogeneous computing architecture of multiple general purpose processors, which may be, for example, a combination of at least two of a CPU, GPU, microprocessor, DSP, ASIC, FPGA; as one example, the processor 1701 may be a CPU + GPU or a CPU + ASIC or a CPU + FPGA.

The communication link 1702 may include a path that conveys information between the aforementioned components.

Communication interface 1704, may be implemented using any transceiver or the like for communicating with other devices or communication networks, such as an ethernet, RAN, wireless Local Area Networks (WLAN), etc.

The memory 1703 may be, but is not limited to, a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a Random Access Memory (RAM) or other type of dynamic storage device that can store information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compact disc, laser disc, optical disc, digital versatile disc, blu-ray disc, etc.), magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. The memory may be self-contained and coupled to the processor through the communication line 1702. The memory may also be integrated with the processor. The memory provided by the embodiment of the application can be generally nonvolatile. The memory 1703 is used for storing computer-executable instructions for implementing the present invention, and is controlled by the processor 1701. The processor 1701 is configured to execute computer-executable instructions stored in the memory 1703 to implement the methods provided in the above-described embodiments of the present application; illustratively, the steps of the methods illustrated in fig. 2, 5 or 6 described above may be implemented.

Optionally, the computer-executable instructions in this embodiment may also be referred to as application program codes, which is not specifically limited in this embodiment.

Illustratively, the processor 1701 may include one or more CPUs, e.g., CPU0 in fig. 12; processor 1701 may also include a CPU and any one of a GPU, ASIC, FPGA, e.g., CPU0+ GPU0 or CPU0+ ASIC0 or CPU0+ FPGA0 of FIG. 12.

Illustratively, the electronic device may include multiple processors, such as the processor 1701 and the processor 1707 in fig. 12. Each of these processors may be a single-core (single-CPU) processor, a multi-core (multi-CPU) processor, or a heterogeneous computing architecture that includes multiple general-purpose processors. A processor herein may refer to one or more devices, circuits, and/or processing cores for processing data (e.g., computer program instructions).

In particular implementations, the electronic device may also include an output device 1705 and an input device 1706, as an example. The output device 1705 is in communication with the processor 1701 and may display information in a variety of ways. For example, the output device 1705 may be a Liquid Crystal Display (LCD), a Light Emitting Diode (LED) display device, a Cathode Ray Tube (CRT) display device, a projector (projector), or the like, and may be, for example, an in-vehicle HUD, an AR-HUD, a display, or the like. The input device 1706, which is in communication with the processor 1701, may receive input from a user in a variety of ways. For example, the input device 1706 may be a mouse, keyboard, touch screen device, or sensing device, among others.

Embodiments of the present application provide a computer-readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the method in the above-described embodiments. Illustratively, the steps of the methods illustrated in fig. 2, 5 or 6 described above may be implemented.

Embodiments of the present application provide a computer program product, which may comprise, for example, computer readable code or a non-transitory computer readable storage medium carrying computer readable code; when the computer program product is run on a computer, it causes the computer to perform the method in the above-described embodiments. Illustratively, the steps of the methods illustrated in fig. 2, 5 or 6 may be described above.

The computer-readable storage medium may be a tangible device that can hold and store the instructions for use by the instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, semiconductor memory device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a Static Random Access Memory (SRAM), a portable compact disc read-only memory (CD-ROM), a Digital Versatile Disc (DVD), a memory stick, a floppy disk, a mechanical coding device, such as a punch card or an in-groove protruding structure with instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media as used herein is not to be interpreted as a transitory signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., optical pulses through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a respective computing/processing device, or to an external computer or external storage device via a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. The network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in the respective computing/processing device.

Computer program instructions for carrying out operations of the present application may be assembler instructions, instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, aspects of the present application are implemented by personalizing an electronic circuit, such as a programmable logic circuit, a Field Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA), with state information of computer-readable program instructions, which can execute the computer-readable program instructions.

Various aspects of the present application are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.

These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer-readable program instructions may also be stored in a computer-readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer-readable medium storing the instructions comprises an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.

The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

Having described embodiments of the present application, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or technical improvements to the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for forecasting an airline ticket sales order, the method comprising:

obtaining air ticket order historical data;

obtaining target prediction data according to the historical target data and at least one characteristic; the historical target data comprises first order quantity information of the air ticket sales orders in a first time period in the air ticket order historical data; the at least one characteristic is constructed according to the travel information of the air ticket sales order in the air ticket order historical data; the target forecast data comprises second order quantity information of the air ticket sales orders in a specified time period in at least one airline department commission-returning protocol;

determining whether to recommend a protocol flight to a user according to the target prediction data; wherein the agreement flight comprises a flight corresponding to the at least one airline commission-returning agreement.

2. The method of claim 1, further comprising:

obtaining a recommendation index of at least one agreement flight according to the user portrait and the agreement flight portrait under the condition of determining that the agreement flight is recommended to the user; the user portrait is constructed according to historical order information of a user; the protocol flight portrait is constructed according to the information of the protocol flight;

and recommending a protocol flight to the user according to the recommendation index.

3. The method of claim 1 or 2, wherein deriving target prediction data from historical target data and at least one characteristic comprises:

obtaining a historical time sequence according to the historical target data;

inputting the historical time sequence and the at least one characteristic into a preset model to obtain the target prediction data;

and the preset model is obtained by training based on an STL time series decomposition algorithm.

4. The method of claim 3, wherein the predetermined model comprises a first submodel, a second submodel, a third submodel; the first sub-model, the second sub-model and the third sub-model are obtained by training based on a regression algorithm;

inputting the historical time sequence and the at least one feature into a preset model to obtain the target prediction data, wherein the step of obtaining the target prediction data comprises the following steps:

obtaining an input sequence according to the historical time sequence and the at least one characteristic;

acquiring a trend item sequence in the (k-1) th iteration, and acquiring a first intermediate sequence according to the input sequence and the trend item sequence in the (k-1) th iteration; wherein the initial value of k is 1, and the initial value of the trend item sequence is 0;

inputting the first intermediate sequence into the first submodel to obtain a second intermediate sequence;

obtaining a third intermediate sequence according to the second intermediate sequence;

inputting the third intermediate sequence into the second submodel to obtain a fourth intermediate sequence;

obtaining a periodic item sequence and a fifth intermediate sequence in the kth iteration according to the input sequence, the second intermediate sequence and the fourth intermediate sequence;

inputting the fifth intermediate sequence into the third submodel to obtain a trend term sequence in the kth iteration;

judging whether the period item sequence in the kth iteration and the trend item sequence in the kth iteration converge, stopping the iteration under the condition that the period item sequence in the kth iteration and the trend item sequence in the kth iteration converge, determining the period item sequence in the latest iteration as the period item sequence corresponding to the input sequence, and determining the trend item sequence in the latest iteration as the trend item sequence corresponding to the input sequence; otherwise, adding 1 to the value of k, and returning to execute the step of acquiring the trend item sequence in the (k-1) th iteration;

and obtaining the target prediction data according to the period item sequence corresponding to the input sequence and the trend item sequence corresponding to the input sequence.

5. The method of claim 4, wherein obtaining the periodic term sequence and the fifth intermediate sequence in the k-th iteration according to the input sequence, the second intermediate sequence and the fourth intermediate sequence comprises:

subtracting the fourth intermediate sequence from the second intermediate sequence to obtain a periodic term sequence in the k iteration;

and subtracting the periodic term sequence in the kth iteration from the input sequence to obtain a fifth intermediate sequence.

6. The method of any of claims 1-5, wherein the objective prediction data further comprises third order quantity information for a ticket sales order within a second time period corresponding to the at least one airline commission-back protocol;

and the third order quantity information of the air ticket sales order in the second time period corresponding to the at least one airline turnback agreement is used for evaluating the airline turnback agreement in the second time period.

7. The method of any of claims 1-6, wherein the first order volume information and the second order volume information include one or more of a total fare, a leg volume, a commission return amount.

8. The method of any one of claims 1-7, wherein the trip information includes one or more of date information, holiday information, city information, business application information.

9. An air ticket sales order prediction apparatus, the apparatus comprising: the acquisition module is used for acquiring air ticket order historical data; the prediction module is used for obtaining target prediction data according to the historical target data and at least one characteristic; the historical target data comprises first order quantity information of the air ticket sales orders in a first time period in the air ticket order historical data; the at least one characteristic is constructed according to the travel information of the air ticket sales order in the air ticket order historical data; the target forecast data comprises second order quantity information of the air ticket sales orders in a specified time period in at least one airline department commission-returning protocol; the recommendation module is used for determining whether to recommend a protocol flight to the user according to the target prediction data; wherein the agreement flight comprises a flight corresponding to the at least one airline commission-returning agreement.

10. The apparatus of claim 9, wherein the recommendation module is further configured to: under the condition that the protocol flight is determined to be recommended to the user, obtaining a recommendation index of at least one protocol flight according to the user portrait and the protocol flight portrait; the user portrait is constructed according to historical order information of a user; the agreement flight portrait is constructed according to the information of the agreement flight; and recommending agreement flights to the user according to the recommendation index.

11. The apparatus of claim 9 or 10, wherein the prediction module is further configured to: obtaining a historical time sequence according to the historical target data; inputting the historical time sequence and the at least one characteristic into a preset model to obtain the target prediction data; the preset model is obtained by training based on an STL time sequence decomposition algorithm.

12. The apparatus of claim 11, wherein the predetermined model comprises a first submodel, a second submodel, a third submodel; the first sub-model, the second sub-model and the third sub-model are obtained by training based on a regression algorithm; the prediction module is further configured to: obtaining an input sequence according to the historical time sequence and the at least one characteristic; acquiring a trend item sequence in the (k-1) th iteration, and acquiring a first intermediate sequence according to the input sequence and the trend item sequence in the (k-1) th iteration; wherein the initial value of k is 1, and the initial value of the trend item sequence is 0; inputting the first intermediate sequence into the first submodel to obtain a second intermediate sequence; obtaining a third intermediate sequence according to the second intermediate sequence; inputting the third intermediate sequence into the second submodel to obtain a fourth intermediate sequence; obtaining a periodic item sequence and a fifth intermediate sequence in the kth iteration according to the input sequence, the second intermediate sequence and the fourth intermediate sequence; inputting the fifth intermediate sequence into the third submodel to obtain a trend term sequence in the kth iteration; judging whether the period item sequence in the kth iteration and the trend item sequence in the kth iteration converge, stopping the iteration under the condition that the period item sequence in the kth iteration and the trend item sequence in the kth iteration converge, determining the period item sequence in the latest iteration as the period item sequence corresponding to the input sequence, and determining the trend item sequence in the latest iteration as the trend item sequence corresponding to the input sequence; otherwise, adding 1 to the value of k, and returning to execute the step of acquiring the trend item sequence in the (k-1) th iteration; and obtaining the target prediction data according to the period item sequence corresponding to the input sequence and the trend item sequence corresponding to the input sequence.

13. The apparatus of claim 12, wherein the prediction module is further configured to: subtracting the fourth intermediate sequence from the second intermediate sequence to obtain a periodic term sequence in the kth iteration; and subtracting the periodic item sequence in the k iteration from the input sequence to obtain the fifth intermediate sequence.

14. The apparatus of any of claims 9-13, wherein the objective prediction data further comprises third order quantity information for a ticket sales order within a second time period corresponding to the at least one airline commission-repay protocol; and the third order quantity information of the air ticket sales order in the second time period corresponding to the at least one airline turnback agreement is used for evaluating the airline turnback agreement in the second time period.

15. The apparatus of any of claims 9-14, wherein the first order amount information and the second order amount information comprise one or more of a total fare, a leg amount, a commission return amount.

16. The apparatus of any one of claims 9-15, wherein the trip information comprises one or more of date information, holiday information, city information, business application information.

17. An electronic device, comprising:

a processor;

a memory for storing processor-executable instructions;

wherein the processor is configured to implement the method of any one of claims 1-8 when executing the instructions.

18. A computer-readable storage medium on which computer program instructions are stored, the computer program instructions, when executed by a processor, implementing the method of any one of claims 1-8.